- Email: [email protected]
Preparation and in vitro characterization of poly(acrylic acid)–cysteine microparticles
Journal of Controlled Release 93 (2003) 29 – 38 www.elsevier.com/locate/jconrel
Preparation and in vitro characterization of poly(acrylic acid)–cysteine microparticles Andreas Bernkop-Schnu¨rch *, Claudia Egger, M. Elhassan Imam, Alexander H. Krauland Institute of Pharmaceutical Technology and Biopharmaceutics, Center of Pharmacy, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria Received 4 April 2003; accepted 8 July 2003
Abstract The purpose of the present study was to prepare and characterize a novel mucoadhesive microparticulate drug delivery system. Microparticles were prepared by the solvent evaporation emulsion technique using a poly(acrylic acid) – cysteine conjugate of an average molecular mass of 450 kDa with an amount of 308 Amol thiol groups per gram polymer. The crosslinking of thiol groups via the formation of disulfide bonds during this preparation process was pH-controlled. The resulting microparticles were characterized with regard to the degree of cross-linking and the amount of remaining free thiol groups, shape, size distribution and stability. Furthermore, the drug release behaviour using bromelain as model drug and the mucoadhesive properties were evaluated. Results demonstrated that the higher the pH of the aqueous phase was during the preparation process, the higher was the degree of cross-linking within the particles. However, even at pH 9, 8.9 F 2.2% of free thiol groups remained on the microparticles. Particles were of spherical and partially porous structure and had a main size in the range of 20 – 60 Am with a center at 35 Am. Because of the formation of disulfide bonds within the particles, they did not disintegrate under physiological conditions within 48 h. In addition, a controlled drug release of bromelain was achieved. Due to the immobilization of thiol groups on poly(acrylic acid), the mucoadhesive properties of the corresponding microparticles were improved threefold. These features should render poly(acrylic acid) – cysteine conjugate microparticles useful as drug delivery system providing a prolonged residence time on mucosal epithelia. D 2003 Elsevier B.V. All rights reserved. Keywords: Poly(acrylic acid) – cysteine conjugate; Microparticles; Mucoadhesion; Thiomers; Cohesion; Cross-linking
1. Introduction The development of mucoadhesive drug delivery systems has received considerable interest in recent * Corresponding author. Tel.: +43-1-4277-55413; fax: +43-14277-9554. E-mail address: [email protected] (A. Bernkop-Schnu¨rch). 0168-3659/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0168-3659(03)00339-0
years, as these systems might be the key to overcome oral bioavailability problems associated with various categories of therapeutic agents such as peptides [1], antisense oligonucleotides [2], heparins [3] or cephalosporines [4]. By intensifying the contact at the absorption site and by increasing the gastrointestinal (GI) residence time of the delivery system, an enhanced drug uptake might be achieved. Although some evidence has been provided for the principal efficacy of
30
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
orally given mucoadhesive delivery systems in humans [5], so far, the concept has certainly not reached its full potential, as the adhesion of drug delivery systems in the GI tract is still insufficient. Even in case of sufficient mucoadhesion, the mucus turnover, which was determined to be in the range of 47– 270 min in rats [6] and 12 – 24 h in humans [7,8], represents an overall limitation for the concept. Attempts to improve the mucoadhesive properties of delivery systems include approaches such as the use of linear poly(ethylene glycol) as adhesion promoter for mucoadhesive hydrogels [9] or the use of lectins or bacterial adhesins [10,11]. Recently, another strategy has been introduced into the pharmaceutical literature. The most commonly bridging structure in biological systems, the disulfide bond, is thereby used to improve adhesion of polymeric carrier systems to mucosal membranes. Thiolated polymers, or the so-called thiomers, are believed to interact with cysteine-rich subdomains of mucus glycoproteins [12] forming disulfide bonds between the mucoadhesive polymer and the mucus layer. Among such thiolated polymers, poly(acrylic acid) – cysteine conjugates turned out to be the most adhesive hydrophilic polymers. Due to the immobilization of thiol groups on poly(acrylic acid), which is even in its unmodified form among the most adhesive polymers [13], the mucoadhesive properties were 20-fold improved [14]. Because of these strong mucoadhesive properties, poly(acrylic acid) – cysteine conjugate was chosen to be used as model polymer within this study. On the other hand, micro- and nanoparticles can also lead to a prolonged GI residence time. By diffusing into the mucus gel layer, their transit time is often significantly prolonged even without exhibiting any mucoadhesive properties [15]. In order to combine both promising strategies— the use of thiolated polymers on the one hand and microparticulate formulations on the other hand—it was the aim of this study to prepare and characterize microparticles being based on a thiolated polymer. Microparticles were prepared via an emulsification solvent evaporation technique. During this process, particles were stabilized by a pH-controlled crosslinking process via the formation of disulfide bonds within the polymeric network. Resulting particles were evaluated in vitro regarding their size, shape, stability, drug release behaviour and mucoadhesive properties.
2. Materials and methods 2.1. Polymer syntheses The poly(acrylic acid) – cysteine conjugate was synthesized according to a method described previously [14]. In brief, 500 mg of cysteine were added to 500 mg poly(acrylic acid) (average molecular mass: 450 kDa; Sigma, St. Louis, MO) hydrated in demineralised water. Cysteine was covalently linked to poly(acrylic acid) via the formation of amide bonds between the primary amino group of cysteine and a carboxylic acid group of the polymer. The reaction was mediated by a water-soluble carbodiimide. After purification of the conjugate via dialysis, the amount of remaining unbound free cysteine was determined with 2,4,6-trinitrobenzenesulfonic acid (TNBS reagent; Sigma) [16]. Fluorescence-labelled poly(acrylic acid) was generated in a similar way. First, 1 g of polycarbophil (Noveon, Raubling, Germany) was hydrated in 100 ml of demineralised water and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDAC; Sigma) was added in a final concentration of 40 mM. The reaction mixture was incubated for 20 min at room temperature in order to activate the carboxylic acid moieties of the polymer. Then, 20 mg of 6amino-fluorescein (Sigma) were dissolved in 2 ml of 1 M NaOH and added to the polymer solution. The reaction was allowed to proceed at pH 5 for 3 h at room temperature. To remove the unbound 6-aminofluorescein, the polymer was dialysed in a dialysis volume of 4.5 l first against 0.2 mM HCl, two times against the same medium but containing 1% NaCl, then four times exhaustively against 0.2 mM HCl and, finally, one time against demineralised water, in each case over 8 h. After dialysis, the aqueous polymer solution was freeze-dried ( 30 jC, 0.1 mbar; Christ Beta 1– 8; Germany) and the conjugate was stored at 4 jC until further use. 2.2. Determination of the thiol/disulfide content The amount of thiol groups on the poly(acrylic acid) – cysteine conjugate and within the resulting microparticles was determined via Ellman’s reagent (5,5V-dithiobis(nitrobenzoic acid)) as described previously [16]. Disulfide content was measured after
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
reduction with NaBH4 and addition of 5,5V-dithiobis(nitrobenzoic acid) as described by Habeeb [17]. 2.3. Preparation of poly(acrylic acid) – cysteine microparticles Microspheres were prepared by a water-in-oil (w/o) emulsification solvent evaporation technique. Each of the poly(acrylic acid) – cysteine conjugate and the corresponding unmodified polymer, both optionally containing 30% of the fluorescence-labelled polycarbophil or 10% of bromelain, were hydrated in demineralised water. The pH was adjusted to 3, 4, 6 and 9 by the addition of 1 M HCl or 1 M NaOH, respectively. Demineralized water was added to obtain a final polymer concentration of 0.75% (m/m). Then, 10 ml of this polymer solution (aqueous phase) were added dropwise into 100 g of paraffin oil (viscosity: 11 – 230 mPa s) (oil phase) containing 0.25% (m/m) Span 20 as emulsifying agent. During this process, the aqueous phase was emulsified into the oil phase by utilizing an ultraturax (Omni 5000; Omni International). Once the emulsion was formed, the dispersed aqueous phase was completely evaporated by maintaining the temperature at 37 jC, bubbling the emulsion with air (5 l/min) and stirring it with a paddle at 300 rpm for 8 h. Petroleum ether (20 ml) was added, and the evaporation process was continued for additional 15 min. After the aqueous phase had been removed, the microspheres were separated from the oil phase by centrifugation (Sorvall RC; 3000 rpm; 5 min), washed several times with petrolether to remove remaining traces of paraffin oil and freeze dried at 30 jC and 0.1 mbar (Christ Beta 1 –8).
31
gold layer (Sputter coater AGAR B7340, Stansted, UK) and viewed in a scanning electron microscope (Philips XL20, Eindhoven, NL). 2.6. Decrease in thiol groups on microparticles The decrease in thiol groups on poly(acrylic acid) – cysteine conjugate microparticles was determined in 50 mM phosphate buffer, pH 6.0. Microparticles (0.5 mg) were suspended in 250 Al buffer solution and incubated at 37 jC under continuous shaking. At predetermined time points, samples were withdrawn and analysed for thiol/disulfide content as described above. 2.7. Dissolution studies The stability of microparticles was analysed in 50 mM phosphate buffer, pH 6.0, at 37 jC. 2.8. Evaluation of the swelling behaviour The water-absorbing capacity was determined by a gravimetric method. First, 0.5 mg of poly(acrylic acid) –cysteine microparticles were incubated in 200 Al of 100 mM phosphate buffer, pH 6.0, preequilibrated to 37 jC. After 1 h of incubation under continuous shaking on an oscillating water bath (GFL 1092; 100 rpm) at 37 jC, the samples were centrifuged for 5 min at 24,000 g and the supernatant was removed. The weight of remaining swollen microparticles was determined and the water uptake was calculated. 2.9. Drug load and release studies
2.4. Particle size determination The particle size of the microspheres was determined in paraffin oil (viscosity: 25– 80 mPa s) as a nondissolving dispersion medium using a laser diffraction particles size analyser (Shimadzu SALD 1100). Particles were suspended by sonification and magnetic stirring during the measurement. 2.5. Scanning electron microscopy The microparticles were dried in a vacuum chamber to remove residual water, sputter-coated with a
Drug load of microspheres was determined as follows: 1 mg of microspheres were hydrated in 1 ml of 0.2 M NaOH and incubated at 37 jC under continuous stirring for 45 min. Samples were centrifuged for 2 min at 24,000g and, bromelain was assayed in the supernatant fluid by measuring the absorbance photometrically (U-3000 Spectrophotometer, Hitachi) at 273 nm. Drug load was calculated by interpolation from a standard curve. The release rate from microparticles containing bromelain as model drug was analysed in vitro. First, 20 mg of microparticles were placed in an Eppendorf
32
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
vessel containing 2.0 ml of release medium (100 mM phosphate buffer, pH 6.0). The vessel was closed, placed on an oscillating water bath (GFL 1092; 100 rpm) and incubated at 37 jC. At predetermined time points, aliquots of 200 Al were withdrawn and replaced with an equal volume of release medium preequilibrated to temperature. Released bromelain was assayed by measuring the absorbance photometrically (U-3000 Spectrophotometer, Hitachi) at 273 nm. Concentrations were calculated by interpolation from a standard curve. 2.10. Mucoadhesion studies Mucoadhesion studies of thiolated and unmodified polymers were performed on freshly excised porcine intestinal mucosa according to a method described previously [18]. In order to be able to observe and quantify microspheres remaining on the mucosa as a function of time, particles comprising a 30% proportion of fluorescence-labelled polycarbophil were used. First, 4 mg of microparticles were hydrated in 140 Al of 100 mM phosphate buffer, pH 6.0. The suspension was then transferred on the porcine mucosa, which was mounted on a platform placed at an angle of 45j. The mucosa was incubated at 37 jC and continuously rinsed with 100 mM phosphate buffer, pH 6.0 (3 ml/h). The amount of adherent particles was investigated visually utilizing a UV lamp. For quantification, the mucus gel layer and the adherent microparticles were scraped off the mucosa at predetermined time points and diluted with the same volume of a mucolytic active solution (100 mM phosphate buffer, pH 6.0, containing 5% trypsin and 5% N-acetylcysteine). Samples were incubated for one hour while shaking at 37 jC. The fluorescence of each sample was then measured at an excitation wavelength of 485 nm and an emission wavelength of 535 nm (Biolise; Spectrafluor; Austria). The amount of adherent particles was calculated by interpolation from a standard curve, which was established from increasing amounts of microparticles being applied to the mucosa and being immediately thereafter scraped off and analysed as described above. 2.11. Statistical data analysis Statistical data analyses were performed using the Student’s t-test, with p < 0.05 as the minimal level of
significance. Calculations were done using the software XIstat version 5.0 (b8.3).
3. Results 3.1. Chemical characterization of polymers The amount of free thiol groups and disulfide bonds distributed all over the polymeric network of the poly(acrylic acid) – cysteine conjugate was quantified via Ellman’s reagent. A total of 308 Amol free thiol groups and 31 Amol disulfide bonds per gram polymer were determined. The amount of unbound free cysteine was determined to be 4 Amol per gram polymer. According to this result, in average, approximately each 450th acrylic acid subunit has been cross-linked via a disulfide bridge. The features of the polymer such as mucoadhesive properties, swelling behaviour and cohesiveness were in good agreement with previously published results [14]. 3.2. Degree of cross-linking during the production process To date, a sufficient stability of polyacrylate microparticles was gained by the addition of crosslinkers such as polyethyleneglycol(400)-dimethacrylate to the polymerisation process, which was performed during the solvent evaporation process of the emulsion [19]. In this study, the stability of polyacrylate microparticles should be guaranteed by the formation of disulfide bonds during the preparation process. Disulfide bond formation can be controlled either by the access of oxygen or by the pH value during the production process. Orientating studies showed that a more accurate control is given by varying the pH value of the aqueous phase and keeping a constant access of oxygen (data not shown). The formation of disulfide bonds depends thereby strongly on the concentration of thiolate anions representing the reactive form of thiol groups for an oxidation process. As the concentration of thiolate anions depends on the pH value, the degree of cross-linking within the microparticles is controlled by the pH of the aqueous phase during the preparation process. The higher the pH, the more disulfide bonds are formed. In order to confirm this
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
33
theory, microparticles were prepared at pH 3, 6 and 9. The degree of cross-linking at these pH values is illustrated in Fig. 1 demonstrating a good correlation between the pH and the resulting content of disulfide bonds. The results of this study showed also that even at a production at pH 9, a sufficiently high concentration of thiol groups, which are essential for high mucoadhesive properties [14], remains available. 3.3. Particle size and shape Analyses of microparticles revealed the same size distribution for microparticles consisting of poly(acrylic acid) – cysteine conjugates and for microparticles consisting of unmodified poly(acrylic acid). In contrast, the particle size was strongly dependent on the polymer concentration in the aqueous phase during the production process. The higher the polymer concentration was, the higher was the average diameter of the resulting microparticles (data not shown). Results of size distribution studies showed furthermore that the
Fig. 2. Size distribution of poly(acrylic acid) – cysteine microparticles prepared via the water-in-oil (w/o) emulsification solvent evaporation technique. The concentration of the polymer in the aqueous phase was 0.75% (m/m). Indicated values are means F S.D. (n = 9).
particle diameter is independent from the pH value of the aqueous phase during the preparation process. In Fig. 2, the size distribution of poly(acrylic acid) – cysteine conjugate microparticles generated from a 0.75% (m/m) aqueous polymer phase is shown. More than 65% of all particles were in the range between 20 and 60 Am, with a center at 35 Am. The size was also confirmed by scanning electron microscopy (Figs. 3 and 4) showing a spherical and partially porous structure of the microparticles. 3.4. Stability of particles under simulated physiological conditions
Fig. 1. Formation of disulfide bonds within microparticles during the production process at pH 3, 6 and 9. Grey bars represent the proportion of free thiol groups. White bars represent the proportion of oxidized thiol groups, i.e. disulfide bonds. Indicated values are means of three experiments F S.D.
The storage stability of poly(acrylic acid) – cysteine conjugates in dry form was shown recently [20]. Within this study, we focused, therefore, only on the chemical and physical stability of the conjugate under simulated physiological conditions. Accordingly, polyacrylate –cysteine microparticles were incubated with 50 mM phosphate buffer, pH 6.0, at 37 jC. Results demonstrated a limited stability of thiol groups on the microparticles. As shown in Fig. 5, the content of thiol groups decreased by almost
34
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 3. SEM micrograph of poly(acrylic acid) – cysteine microparticles. Bar represents 10 Am.
75% within 3 h of incubation. Hence, the thiol groups on microparticles are highly reactive and are capable of forming disulfide bonds not only within the particles but also with thiol groups of the mucus gel layer. As the adhesion of microparticles within the small intestine has to take place within minutes, the velocity of this oxidation process does not seem to be too rapid. Dissolution studies demonstrated that poly(acrylic acid) – cysteine microparticles do not dissolve or erode within an incubation period of 48 h, whereas microparticles comprising the unmodified polymer eroded within minutes.
Fig. 4. SEM micrograph of poly(acrylic acid) – cysteine microparticles. Bar represents 20 Am.
Fig. 5. Oxidation of thiol groups on microparticles under simulated physiological conditions (50 mM phosphate buffer, pH 6.0). Indicated values are means of three experiments F S.D.
3.5. Evaluation of the swelling behaviour Poly(acrylic acid) –cysteine microparticles swelled within minutes. Results of swelling studies showed a 29.3 F 3.4-fold increase in weight of completely hydrated microparticles. 3.6. Drug load and release studies As model drug for release studies bromelain was chosen. The peptide drug is currently on the market in form of tablets for treatment of inflammatory diseases (e.g. Phlogenzym tablets, Mucos, Germany). The oral bioavailability of this therapeutic peptide, however, is only in the range of a few percent [21]. The use of mucoadhesive microparticles as drug delivery system might strongly improve its oral bioavailability. In order to avoid any unintended thiol/disulfide exchange reactions between bromelain and the thiolated polymer, poly(acrylic acid) – cysteine microparticles were prepared at pH 4. The drug load was determined to be 98.1%. Drug release studies out of these particles demonstrated a controlled drug release over a period of 1 h. The results are shown in Fig. 6. A more sustained release might be achieved by the addition of more hydrophobic polymers such as polymethacrylatemethylesters to the microparticles guaranteeing a compar-
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 6. Release profile of bromelain from poly(acrylic acid) – cysteine microspheres prepared at pH 4. The release study was performed in 100 mM phosphate buffer, pH 6.0, at 37 jC. Indicated values are means of three experiments F S.D.
35
particles displayed threefold higher mucoadhesive properties. A reason for this observation might be, on the one hand, the capability of thiolated microparticles to form covalent bonds via disulfide bridges with the mucus gel layer [12]. According to the chemical structure of the thiolated polymer, 0.16 – 1.58% thiolate anions being responsible for the formation of new disulfide bonds should be available at pH 6 –7. Calculations were performed with the ACD software (Toronto, Canada). On the other hand, the cohesive properties of the thiolated microparticles— being also essential for high mucoadhesive properties—are comparatively higher. Thiolated microparticles remained very stable on the mucosa, whereas unmodified poly(acrylic acid) microparticles disintegrated within minutes. In Fig. 8, fluorescence-labelled poly(acrylic acid) and poly(acrylic acid) – cysteine microparticles are shown a few minutes and 3 h after application on the rinsed mucosa demonstrating the disintegration of the unmodified particles. Even if the adhesion of the particles to the mucus gel layer is high, the delivery system will not remain on the mucosa
atively slower swelling process and subsequently drug release behaviour. 3.7. Mucoadhesion studies The small intestinal transit time for multiple-unit dosage forms in humans was determined to be in the range of 3 h [15]. The adhesion of microparticles on freshly excised small intestinal mucosa was, therefore, investigated within this time period. When applied in dry form, both poly(acrylic acid) microparticles and poly(acrylic acid) – cysteine microparticles remained entirely on the mucosa during the observation period. In order to avoid mucoadhesion of the particles by this simple ‘adhesion by hydration process’ [22], which will not take place in the small intestine, microparticles were applied to the mucosa after having been fully hydrated in a physiological buffer solution. Results of these mucoadhesion studies as shown in Fig. 7 demonstrated that a certain fraction of applied particles was rapidly washed out. The rest, however, showed significant mucoadhesive properties. The mucoadhesion of particles comprising thiolated poly(acrylic acid) was significantly higher than that of particles being based on the corresponding unmodified polymer. Thiolated
Fig. 7. Adhesion of poly(acrylic acid) – cysteine microparticles ( ) and poly(acrylic acid) microparticles (o) on freshly excised small intestinal mucosa rinsed with 100 mM phosphate buffer, pH 6.0, at 37 jC as function of time. Indicated values are means ( F S.D.) of three to six experiments.
y
36
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 8. Images of fluorescence-labelled poly(acrylic acid) – cysteine and poly(acrylic acid) microparticles on freshly excised small intestinal mucosa rinsed with 100 mM phosphate buffer, pH 6.0, at 37 jC a few minutes and 3 h after application. Left top: poly(acrylic acid) – cysteine microparticles, t = 5 min. Left bottom: poly(acrylic acid) – cysteine microparticles, t = 180 min. Right top: poly(acrylic acid) microparticles, t = 5 min. Right bottom: poly(acrylic acid) microparticles, t = 180 min.
when the adhesive bond fails within the mucoadhesive polymer itself [23].
4. Discussion Within this study, a model drug whose oral bioavailability should be improved by the use of mucoadhesive microparticles was used. The potential of orally given mucoadhesive microparticles
could meanwhile be verified by a study in human volunteers demonstrating a strongly improved oral bioavailability of riboflavin when the drug was incorporated in mucoadhesive microparticles [5]. Similar results might be achieved for bromelain being incorporated in poly(acrylic acid) – cysteine microparticles. Apart from an oral use of poly(acrylic acid) – cysteine microparticles, they might also be useful in order to improve the bioavailability of various poorly absorbed drugs being administered
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
via other transmucosal routes such as the nasal or vaginal mucosa. Vidgren et al. [24], for instance, could show in human volunteers a significantly prolonged residence time on the nasal mucosa for microparticles being based on unmodified poly(acrylic acid) in comparison to nonadhesive particles. As the mucoadhesive properties of poly(acrylic acid) microparticles could be strongly improved by the immobilization of thiol groups, a more pronounced effect as described by Vidgren et al. can be expected but has to be verified by according in vivo studies. Apart from the high cohesive and mucoadhesive properties, poly(acrylic acid) – cysteine microspheres should also lead to an improved drug uptake. Due to the intimate contact of adherent microparticles with the absorption membrane, a comparatively steep concentration gradient of the drug to the epithelium can be achieved, representing the driving force for a passive absorption. In addition, poly(acrylic acid) exhibits per se a permeation enhancing effect [25]. Moreover, Clausen and Bernkop-Schnu¨rch [26] could demonstrate that due to the immobilization of thiol groups on poly(acrylic acid), its permeation enhancing effect can be significantly improved. According to these results also, poly(acrylic acid) – cysteine microspheres should exhibit strong permeation enhancing properties. Recently, it was shown that this permeation enhancing effect of thiolated polymers can even be further improved by the addition of glutathione, which seems to mediate the permeation-enhancing effect of thiomers [27]. Hence, the addition of glutathione to poly(acrylic acid) – cysteine microspheres might be advantageous when an improved mucosal uptake of the incorporated drug is desired. In case of peptide and protein delivery, the enzyme inhibitory properties of poly(acrylic acid) should also contribute to an improved drug uptake. Poly(acrylic acid) displays a strong inhibitory effect towards the endopeptidase trypsin [28]. By the immobilization of thiol groups on poly(acrylic acid) also, a pronounced inhibitory effect towards exopeptidases such as carboxypeptidase A and B or aminopeptidase N can be achieved [29,30]. Being incorporated in poly(acrylic acid) – cysteine microparticles, therapeutic peptides or proteins such as bromelain should, therefore, be protected towards an enzymatic attack of these enzymes.
37
5. Conclusion Poly(acrylic acid) – cysteine microparticles display high cohesive properties due to a cross-linking process via the formation of disulfide bonds during preparation (I). Remaining free thiol groups seem to be responsible for relatively high mucoadhesive properties (II). Furthermore, a controlled drug release might be guaranteed by these particles (III) and— based on the results of prior studies performed with poly(acrylic acid) – cysteine—a permeation enhancing effect (IV) and enzyme inhibitory properties (V) can be expected. Because of these features, poly(acrylic acid) – cysteine microparticles seem to represent a promising novel delivery system for therapeutic agents whose uptake from the mucosal membranes such as the intestinal, nasal or vaginal should be prolonged and/or enhanced. Acknowledgements This work was supported by grant no. P13085MOB from the Fonds zur Fo¨rderung der wissenschaftlichen Forschung (FWF) to A. BernkopSchnu¨rch. The authors wish to thank Mr. Stro¨bel et al. from the slaughterhouse Totzenbach for supply of porcine intestinal mucosa. References [1] R.B. Shah, F. Ahsan, M.A. Khan, Oral delivery of proteins: progress and prognostication, Crit. Rev. Ther. Drug Carr. Syst. 19 (2002) 135 – 169. [2] A.A. Raoof, Z. Ramtoola, B. McKenna, R.Z. Yu, G. Hardee, R.S. Geary, Effect of sodium caprate on the intestinal absorption of two modified antisense oligonucleotides in pigs, Eur. J. Pharm. Sci. 17 (2002) 131 – 139. [3] S.R. Money, J.W. York, Development of oral heparin therapy for prophylaxis and treatment of deep venous thrombosis, Cardiovasc. Surg. 9 (2001) 211 – 218. [4] M.E. Klepser, M.N. Marangos, K.B. Patel, D.P. Nicolau, R. Quintiliani, C.H. Nightingale, Clinical pharmacokinetics of newer cephalosporins, Clin. Pharmacokinet. 28 (1995) 361 – 384. [5] Y. Akiyama, N. Nagahara, E. Nara, M. Kitano, S. Iwasa, I. Yamamoto, J. Azuma, Y. Ogawa, Evaluation of oral mucoadhesive microspheres in man on the basis of the pharmacokinetics of furosemide and riboflavin, compounds with limited gastrointestinal absorption sites, J. Pharm. Pharmacol. 50 (1998) 159 – 166.
38
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
[6] C.-M. Lehr, F.G.J. Poelma, H.E. Junginger, An estimate of turnover time of intestinal mucus gel layer in the rat in situ loop, Int. J. Pharm. 70 (1991) 235. [7] J.F. Forstner, Intestinal mucins in health and disease, Digestion 17 (1978) 234. [8] A. Allen, D.A. Hutton, J.P. Pearson, L.A. Sellers, in: J. Nugent, M. O’Connor (Eds.), Mucus and Mucosa, Ciba Foundation Symposium, vol. 109 (1998) p. 1984. [9] J.J. Sahlin, N.A. Peppas, Enhanced hydrogel adhesion by polymer interdiffusion: use of linear poly(ethylene glycol) as an adhesion promoter, J. Biomater. Sci., Polym. Ed. 8 (1997) 421 – 436. [10] B. Naisbett, J. Woodley, The potential use of tomato lectin for oral drug delivery. Lectin binding to rat small intestine in vitro, Int. J. Pharm. 107 (1994) 223 – 230. [11] A. Bernkop-Schnu¨rch, F. Gabor, M.P. Szostak, W. Lubitz, An adhesive drug delivery system based on K99-fimbriae, Eur. J. Pharm. Sci. 3 (1995) 293 – 299. [12] V.M. Leitner, G.F. Walker, A. Bernkop-Schnu¨rch, Thiolated polymers: evidence for the formation of disulphide bonds with mucus glycoproteins, Eur. J. Pharm. Biopharm. (2003) (in press). [13] J.D. Smart, I.W. Kellaway, H.E.C. Worthington, An in vitro investigation of mucosa-adhesive materials for use in controlled drug delivery, J. Pharm. Pharmacol. 36 (1984) 295 – 299. [14] M.K. Marschu¨tz, A. Bernkop-Schnu¨rch, Thiolated polymers: self-crosslinking properties of thiolated 450 kDa poly(acrylic acid) and their influence on mucoadhesion, Eur. J. Pharm. Sci. 15 (2002) 387 – 394. [15] A.J. Coupe, S.S. Davis, I.R. Wilding, Variation in gastrointestinal transit of pharmaceutical dosage forms in healthy subjects, Pharm. Res. 8 (1991) 360 – 364. [16] A. Bernkop-Schnu¨rch, C.E. Kast, M.F. Richter, Improvement in the mucoadhesive properties of alginate by the covalent attachment of cysteine, J. Control. Release 71 (2001) 277 – 285. [17] A.F. Habeeb, A sensitive method for localization of disulfide containing peptides in column effluents, Anal. Biochem. 56 (1973) 60 – 65. [18] K.V. Ranga Rao, P. Buri, A novel in situ method to test polymers and coated microparticles for bioadhesion, Int. J. Pharm. 52 (1989) 265 – 270. [19] B. Kriwet, E. Walter, T. Kissel, Synthesis of bioadhesive
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
poly(acrylic acid) nano- and microparticles using an inverse emulsion polymerization method for the entrapment of hydrophilic drug candidates, J. Control. Release 56 (1998) 149 – 158. A. Bernkop-Schnu¨rch, M.D. Hornof, C.E. Kast, N. Langoth, Thiolated polymers: stability of thiol moieties under different storage conditions, Sci. Pharm. 70 (2002) 331 – 339. A. Bernkop-Schnu¨rch, R. Giovanelli, C. Valenta, Peroral administration of enzymes: strategies to improve the galenic of dosage forms for trypsin and bromelain, Drug Dev. Ind. Pharm. 26 (2000) 115 – 121. C.-M. Lehr, From sticky stuff to sweet receptors—achievements, limits and novel approaches to bioadhesion, Eur. J. Drug Metab. Pharmacokinet. 21 (1996) 139 – 148. A. Bernkop-Schnu¨rch, S. Steininger, Synthesis and characterisation of mucoadhesive thiolated polymers, Int. J. Pharm. 194 (2000) 239 – 247. P. Vidgren, M. Vidgren, P. Vainio, J. Nuutinen, P. Paronen, Double-labelling technique in the evaluation of nasal mucoadhesion of disodium cromoglycate microspheres, Int. J. Pharm. 73 (1991) 131 – 136. C.E. Kast, A. Bernkop-Schnu¨rch, Influence of the molecular mass on the permeation enhancing effect of different poly(acrylates), STP Pharma 12 (2002) 351 – 356. A.E. Clausen, A. Bernkop-Schnu¨rch, In vitro evaluation of the permeation-enhancing effect of thiolated polycarbophil, J. Pharm. Sci. 89 (2000) 1253 – 1261. A.E. Clausen, C.E. Kast, A. Bernkop-Schnu¨rch, The role of glutathione in the permeation enhancing effect of thiolated polymers, Pharm. Res. 19 (2002) 602 – 608. H.L. Luessen, J.C. Verhoef, G. Borchard, C.-M. Lehr, A.G. de Boer, H.E. Junginger, Mucoadhesive polymers in peroral peptide drug delivery: II. Carbomer and polycarbophil are potent inhibitors of the intestinal proteolytic enzyme trypsin, Pharm. Res. 12 (1995) 1293 – 1298. A. Bernkop-Schnu¨rch, H. Zarti, G.F. Walker, Thiolation of polycarbophil enhances its inhibition of soluble and intestinal brush border membrane bound aminopeptidase N, J. Pharm. Sci. 90 (2001) 1907 – 1914. A. Bernkop-Schnu¨rch, S. Thaler, Polycarbophil – cysteine conjugates as platforms for oral (poly)peptide delivery systems, J. Pharm. Sci. 89 (2000) 901 – 909.
Preparation and in vitro characterization of poly(acrylic acid)–cysteine microparticles Andreas Bernkop-Schnu¨rch *, Claudia Egger, M. Elhassan Imam, Alexander H. Krauland Institute of Pharmaceutical Technology and Biopharmaceutics, Center of Pharmacy, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria Received 4 April 2003; accepted 8 July 2003
Abstract The purpose of the present study was to prepare and characterize a novel mucoadhesive microparticulate drug delivery system. Microparticles were prepared by the solvent evaporation emulsion technique using a poly(acrylic acid) – cysteine conjugate of an average molecular mass of 450 kDa with an amount of 308 Amol thiol groups per gram polymer. The crosslinking of thiol groups via the formation of disulfide bonds during this preparation process was pH-controlled. The resulting microparticles were characterized with regard to the degree of cross-linking and the amount of remaining free thiol groups, shape, size distribution and stability. Furthermore, the drug release behaviour using bromelain as model drug and the mucoadhesive properties were evaluated. Results demonstrated that the higher the pH of the aqueous phase was during the preparation process, the higher was the degree of cross-linking within the particles. However, even at pH 9, 8.9 F 2.2% of free thiol groups remained on the microparticles. Particles were of spherical and partially porous structure and had a main size in the range of 20 – 60 Am with a center at 35 Am. Because of the formation of disulfide bonds within the particles, they did not disintegrate under physiological conditions within 48 h. In addition, a controlled drug release of bromelain was achieved. Due to the immobilization of thiol groups on poly(acrylic acid), the mucoadhesive properties of the corresponding microparticles were improved threefold. These features should render poly(acrylic acid) – cysteine conjugate microparticles useful as drug delivery system providing a prolonged residence time on mucosal epithelia. D 2003 Elsevier B.V. All rights reserved. Keywords: Poly(acrylic acid) – cysteine conjugate; Microparticles; Mucoadhesion; Thiomers; Cohesion; Cross-linking
1. Introduction The development of mucoadhesive drug delivery systems has received considerable interest in recent * Corresponding author. Tel.: +43-1-4277-55413; fax: +43-14277-9554. E-mail address: [email protected] (A. Bernkop-Schnu¨rch). 0168-3659/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0168-3659(03)00339-0
years, as these systems might be the key to overcome oral bioavailability problems associated with various categories of therapeutic agents such as peptides [1], antisense oligonucleotides [2], heparins [3] or cephalosporines [4]. By intensifying the contact at the absorption site and by increasing the gastrointestinal (GI) residence time of the delivery system, an enhanced drug uptake might be achieved. Although some evidence has been provided for the principal efficacy of
30
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
orally given mucoadhesive delivery systems in humans [5], so far, the concept has certainly not reached its full potential, as the adhesion of drug delivery systems in the GI tract is still insufficient. Even in case of sufficient mucoadhesion, the mucus turnover, which was determined to be in the range of 47– 270 min in rats [6] and 12 – 24 h in humans [7,8], represents an overall limitation for the concept. Attempts to improve the mucoadhesive properties of delivery systems include approaches such as the use of linear poly(ethylene glycol) as adhesion promoter for mucoadhesive hydrogels [9] or the use of lectins or bacterial adhesins [10,11]. Recently, another strategy has been introduced into the pharmaceutical literature. The most commonly bridging structure in biological systems, the disulfide bond, is thereby used to improve adhesion of polymeric carrier systems to mucosal membranes. Thiolated polymers, or the so-called thiomers, are believed to interact with cysteine-rich subdomains of mucus glycoproteins [12] forming disulfide bonds between the mucoadhesive polymer and the mucus layer. Among such thiolated polymers, poly(acrylic acid) – cysteine conjugates turned out to be the most adhesive hydrophilic polymers. Due to the immobilization of thiol groups on poly(acrylic acid), which is even in its unmodified form among the most adhesive polymers [13], the mucoadhesive properties were 20-fold improved [14]. Because of these strong mucoadhesive properties, poly(acrylic acid) – cysteine conjugate was chosen to be used as model polymer within this study. On the other hand, micro- and nanoparticles can also lead to a prolonged GI residence time. By diffusing into the mucus gel layer, their transit time is often significantly prolonged even without exhibiting any mucoadhesive properties [15]. In order to combine both promising strategies— the use of thiolated polymers on the one hand and microparticulate formulations on the other hand—it was the aim of this study to prepare and characterize microparticles being based on a thiolated polymer. Microparticles were prepared via an emulsification solvent evaporation technique. During this process, particles were stabilized by a pH-controlled crosslinking process via the formation of disulfide bonds within the polymeric network. Resulting particles were evaluated in vitro regarding their size, shape, stability, drug release behaviour and mucoadhesive properties.
2. Materials and methods 2.1. Polymer syntheses The poly(acrylic acid) – cysteine conjugate was synthesized according to a method described previously [14]. In brief, 500 mg of cysteine were added to 500 mg poly(acrylic acid) (average molecular mass: 450 kDa; Sigma, St. Louis, MO) hydrated in demineralised water. Cysteine was covalently linked to poly(acrylic acid) via the formation of amide bonds between the primary amino group of cysteine and a carboxylic acid group of the polymer. The reaction was mediated by a water-soluble carbodiimide. After purification of the conjugate via dialysis, the amount of remaining unbound free cysteine was determined with 2,4,6-trinitrobenzenesulfonic acid (TNBS reagent; Sigma) [16]. Fluorescence-labelled poly(acrylic acid) was generated in a similar way. First, 1 g of polycarbophil (Noveon, Raubling, Germany) was hydrated in 100 ml of demineralised water and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDAC; Sigma) was added in a final concentration of 40 mM. The reaction mixture was incubated for 20 min at room temperature in order to activate the carboxylic acid moieties of the polymer. Then, 20 mg of 6amino-fluorescein (Sigma) were dissolved in 2 ml of 1 M NaOH and added to the polymer solution. The reaction was allowed to proceed at pH 5 for 3 h at room temperature. To remove the unbound 6-aminofluorescein, the polymer was dialysed in a dialysis volume of 4.5 l first against 0.2 mM HCl, two times against the same medium but containing 1% NaCl, then four times exhaustively against 0.2 mM HCl and, finally, one time against demineralised water, in each case over 8 h. After dialysis, the aqueous polymer solution was freeze-dried ( 30 jC, 0.1 mbar; Christ Beta 1– 8; Germany) and the conjugate was stored at 4 jC until further use. 2.2. Determination of the thiol/disulfide content The amount of thiol groups on the poly(acrylic acid) – cysteine conjugate and within the resulting microparticles was determined via Ellman’s reagent (5,5V-dithiobis(nitrobenzoic acid)) as described previously [16]. Disulfide content was measured after
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
reduction with NaBH4 and addition of 5,5V-dithiobis(nitrobenzoic acid) as described by Habeeb [17]. 2.3. Preparation of poly(acrylic acid) – cysteine microparticles Microspheres were prepared by a water-in-oil (w/o) emulsification solvent evaporation technique. Each of the poly(acrylic acid) – cysteine conjugate and the corresponding unmodified polymer, both optionally containing 30% of the fluorescence-labelled polycarbophil or 10% of bromelain, were hydrated in demineralised water. The pH was adjusted to 3, 4, 6 and 9 by the addition of 1 M HCl or 1 M NaOH, respectively. Demineralized water was added to obtain a final polymer concentration of 0.75% (m/m). Then, 10 ml of this polymer solution (aqueous phase) were added dropwise into 100 g of paraffin oil (viscosity: 11 – 230 mPa s) (oil phase) containing 0.25% (m/m) Span 20 as emulsifying agent. During this process, the aqueous phase was emulsified into the oil phase by utilizing an ultraturax (Omni 5000; Omni International). Once the emulsion was formed, the dispersed aqueous phase was completely evaporated by maintaining the temperature at 37 jC, bubbling the emulsion with air (5 l/min) and stirring it with a paddle at 300 rpm for 8 h. Petroleum ether (20 ml) was added, and the evaporation process was continued for additional 15 min. After the aqueous phase had been removed, the microspheres were separated from the oil phase by centrifugation (Sorvall RC; 3000 rpm; 5 min), washed several times with petrolether to remove remaining traces of paraffin oil and freeze dried at 30 jC and 0.1 mbar (Christ Beta 1 –8).
31
gold layer (Sputter coater AGAR B7340, Stansted, UK) and viewed in a scanning electron microscope (Philips XL20, Eindhoven, NL). 2.6. Decrease in thiol groups on microparticles The decrease in thiol groups on poly(acrylic acid) – cysteine conjugate microparticles was determined in 50 mM phosphate buffer, pH 6.0. Microparticles (0.5 mg) were suspended in 250 Al buffer solution and incubated at 37 jC under continuous shaking. At predetermined time points, samples were withdrawn and analysed for thiol/disulfide content as described above. 2.7. Dissolution studies The stability of microparticles was analysed in 50 mM phosphate buffer, pH 6.0, at 37 jC. 2.8. Evaluation of the swelling behaviour The water-absorbing capacity was determined by a gravimetric method. First, 0.5 mg of poly(acrylic acid) –cysteine microparticles were incubated in 200 Al of 100 mM phosphate buffer, pH 6.0, preequilibrated to 37 jC. After 1 h of incubation under continuous shaking on an oscillating water bath (GFL 1092; 100 rpm) at 37 jC, the samples were centrifuged for 5 min at 24,000 g and the supernatant was removed. The weight of remaining swollen microparticles was determined and the water uptake was calculated. 2.9. Drug load and release studies
2.4. Particle size determination The particle size of the microspheres was determined in paraffin oil (viscosity: 25– 80 mPa s) as a nondissolving dispersion medium using a laser diffraction particles size analyser (Shimadzu SALD 1100). Particles were suspended by sonification and magnetic stirring during the measurement. 2.5. Scanning electron microscopy The microparticles were dried in a vacuum chamber to remove residual water, sputter-coated with a
Drug load of microspheres was determined as follows: 1 mg of microspheres were hydrated in 1 ml of 0.2 M NaOH and incubated at 37 jC under continuous stirring for 45 min. Samples were centrifuged for 2 min at 24,000g and, bromelain was assayed in the supernatant fluid by measuring the absorbance photometrically (U-3000 Spectrophotometer, Hitachi) at 273 nm. Drug load was calculated by interpolation from a standard curve. The release rate from microparticles containing bromelain as model drug was analysed in vitro. First, 20 mg of microparticles were placed in an Eppendorf
32
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
vessel containing 2.0 ml of release medium (100 mM phosphate buffer, pH 6.0). The vessel was closed, placed on an oscillating water bath (GFL 1092; 100 rpm) and incubated at 37 jC. At predetermined time points, aliquots of 200 Al were withdrawn and replaced with an equal volume of release medium preequilibrated to temperature. Released bromelain was assayed by measuring the absorbance photometrically (U-3000 Spectrophotometer, Hitachi) at 273 nm. Concentrations were calculated by interpolation from a standard curve. 2.10. Mucoadhesion studies Mucoadhesion studies of thiolated and unmodified polymers were performed on freshly excised porcine intestinal mucosa according to a method described previously [18]. In order to be able to observe and quantify microspheres remaining on the mucosa as a function of time, particles comprising a 30% proportion of fluorescence-labelled polycarbophil were used. First, 4 mg of microparticles were hydrated in 140 Al of 100 mM phosphate buffer, pH 6.0. The suspension was then transferred on the porcine mucosa, which was mounted on a platform placed at an angle of 45j. The mucosa was incubated at 37 jC and continuously rinsed with 100 mM phosphate buffer, pH 6.0 (3 ml/h). The amount of adherent particles was investigated visually utilizing a UV lamp. For quantification, the mucus gel layer and the adherent microparticles were scraped off the mucosa at predetermined time points and diluted with the same volume of a mucolytic active solution (100 mM phosphate buffer, pH 6.0, containing 5% trypsin and 5% N-acetylcysteine). Samples were incubated for one hour while shaking at 37 jC. The fluorescence of each sample was then measured at an excitation wavelength of 485 nm and an emission wavelength of 535 nm (Biolise; Spectrafluor; Austria). The amount of adherent particles was calculated by interpolation from a standard curve, which was established from increasing amounts of microparticles being applied to the mucosa and being immediately thereafter scraped off and analysed as described above. 2.11. Statistical data analysis Statistical data analyses were performed using the Student’s t-test, with p < 0.05 as the minimal level of
significance. Calculations were done using the software XIstat version 5.0 (b8.3).
3. Results 3.1. Chemical characterization of polymers The amount of free thiol groups and disulfide bonds distributed all over the polymeric network of the poly(acrylic acid) – cysteine conjugate was quantified via Ellman’s reagent. A total of 308 Amol free thiol groups and 31 Amol disulfide bonds per gram polymer were determined. The amount of unbound free cysteine was determined to be 4 Amol per gram polymer. According to this result, in average, approximately each 450th acrylic acid subunit has been cross-linked via a disulfide bridge. The features of the polymer such as mucoadhesive properties, swelling behaviour and cohesiveness were in good agreement with previously published results [14]. 3.2. Degree of cross-linking during the production process To date, a sufficient stability of polyacrylate microparticles was gained by the addition of crosslinkers such as polyethyleneglycol(400)-dimethacrylate to the polymerisation process, which was performed during the solvent evaporation process of the emulsion [19]. In this study, the stability of polyacrylate microparticles should be guaranteed by the formation of disulfide bonds during the preparation process. Disulfide bond formation can be controlled either by the access of oxygen or by the pH value during the production process. Orientating studies showed that a more accurate control is given by varying the pH value of the aqueous phase and keeping a constant access of oxygen (data not shown). The formation of disulfide bonds depends thereby strongly on the concentration of thiolate anions representing the reactive form of thiol groups for an oxidation process. As the concentration of thiolate anions depends on the pH value, the degree of cross-linking within the microparticles is controlled by the pH of the aqueous phase during the preparation process. The higher the pH, the more disulfide bonds are formed. In order to confirm this
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
33
theory, microparticles were prepared at pH 3, 6 and 9. The degree of cross-linking at these pH values is illustrated in Fig. 1 demonstrating a good correlation between the pH and the resulting content of disulfide bonds. The results of this study showed also that even at a production at pH 9, a sufficiently high concentration of thiol groups, which are essential for high mucoadhesive properties [14], remains available. 3.3. Particle size and shape Analyses of microparticles revealed the same size distribution for microparticles consisting of poly(acrylic acid) – cysteine conjugates and for microparticles consisting of unmodified poly(acrylic acid). In contrast, the particle size was strongly dependent on the polymer concentration in the aqueous phase during the production process. The higher the polymer concentration was, the higher was the average diameter of the resulting microparticles (data not shown). Results of size distribution studies showed furthermore that the
Fig. 2. Size distribution of poly(acrylic acid) – cysteine microparticles prepared via the water-in-oil (w/o) emulsification solvent evaporation technique. The concentration of the polymer in the aqueous phase was 0.75% (m/m). Indicated values are means F S.D. (n = 9).
particle diameter is independent from the pH value of the aqueous phase during the preparation process. In Fig. 2, the size distribution of poly(acrylic acid) – cysteine conjugate microparticles generated from a 0.75% (m/m) aqueous polymer phase is shown. More than 65% of all particles were in the range between 20 and 60 Am, with a center at 35 Am. The size was also confirmed by scanning electron microscopy (Figs. 3 and 4) showing a spherical and partially porous structure of the microparticles. 3.4. Stability of particles under simulated physiological conditions
Fig. 1. Formation of disulfide bonds within microparticles during the production process at pH 3, 6 and 9. Grey bars represent the proportion of free thiol groups. White bars represent the proportion of oxidized thiol groups, i.e. disulfide bonds. Indicated values are means of three experiments F S.D.
The storage stability of poly(acrylic acid) – cysteine conjugates in dry form was shown recently [20]. Within this study, we focused, therefore, only on the chemical and physical stability of the conjugate under simulated physiological conditions. Accordingly, polyacrylate –cysteine microparticles were incubated with 50 mM phosphate buffer, pH 6.0, at 37 jC. Results demonstrated a limited stability of thiol groups on the microparticles. As shown in Fig. 5, the content of thiol groups decreased by almost
34
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 3. SEM micrograph of poly(acrylic acid) – cysteine microparticles. Bar represents 10 Am.
75% within 3 h of incubation. Hence, the thiol groups on microparticles are highly reactive and are capable of forming disulfide bonds not only within the particles but also with thiol groups of the mucus gel layer. As the adhesion of microparticles within the small intestine has to take place within minutes, the velocity of this oxidation process does not seem to be too rapid. Dissolution studies demonstrated that poly(acrylic acid) – cysteine microparticles do not dissolve or erode within an incubation period of 48 h, whereas microparticles comprising the unmodified polymer eroded within minutes.
Fig. 4. SEM micrograph of poly(acrylic acid) – cysteine microparticles. Bar represents 20 Am.
Fig. 5. Oxidation of thiol groups on microparticles under simulated physiological conditions (50 mM phosphate buffer, pH 6.0). Indicated values are means of three experiments F S.D.
3.5. Evaluation of the swelling behaviour Poly(acrylic acid) –cysteine microparticles swelled within minutes. Results of swelling studies showed a 29.3 F 3.4-fold increase in weight of completely hydrated microparticles. 3.6. Drug load and release studies As model drug for release studies bromelain was chosen. The peptide drug is currently on the market in form of tablets for treatment of inflammatory diseases (e.g. Phlogenzym tablets, Mucos, Germany). The oral bioavailability of this therapeutic peptide, however, is only in the range of a few percent [21]. The use of mucoadhesive microparticles as drug delivery system might strongly improve its oral bioavailability. In order to avoid any unintended thiol/disulfide exchange reactions between bromelain and the thiolated polymer, poly(acrylic acid) – cysteine microparticles were prepared at pH 4. The drug load was determined to be 98.1%. Drug release studies out of these particles demonstrated a controlled drug release over a period of 1 h. The results are shown in Fig. 6. A more sustained release might be achieved by the addition of more hydrophobic polymers such as polymethacrylatemethylesters to the microparticles guaranteeing a compar-
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 6. Release profile of bromelain from poly(acrylic acid) – cysteine microspheres prepared at pH 4. The release study was performed in 100 mM phosphate buffer, pH 6.0, at 37 jC. Indicated values are means of three experiments F S.D.
35
particles displayed threefold higher mucoadhesive properties. A reason for this observation might be, on the one hand, the capability of thiolated microparticles to form covalent bonds via disulfide bridges with the mucus gel layer [12]. According to the chemical structure of the thiolated polymer, 0.16 – 1.58% thiolate anions being responsible for the formation of new disulfide bonds should be available at pH 6 –7. Calculations were performed with the ACD software (Toronto, Canada). On the other hand, the cohesive properties of the thiolated microparticles— being also essential for high mucoadhesive properties—are comparatively higher. Thiolated microparticles remained very stable on the mucosa, whereas unmodified poly(acrylic acid) microparticles disintegrated within minutes. In Fig. 8, fluorescence-labelled poly(acrylic acid) and poly(acrylic acid) – cysteine microparticles are shown a few minutes and 3 h after application on the rinsed mucosa demonstrating the disintegration of the unmodified particles. Even if the adhesion of the particles to the mucus gel layer is high, the delivery system will not remain on the mucosa
atively slower swelling process and subsequently drug release behaviour. 3.7. Mucoadhesion studies The small intestinal transit time for multiple-unit dosage forms in humans was determined to be in the range of 3 h [15]. The adhesion of microparticles on freshly excised small intestinal mucosa was, therefore, investigated within this time period. When applied in dry form, both poly(acrylic acid) microparticles and poly(acrylic acid) – cysteine microparticles remained entirely on the mucosa during the observation period. In order to avoid mucoadhesion of the particles by this simple ‘adhesion by hydration process’ [22], which will not take place in the small intestine, microparticles were applied to the mucosa after having been fully hydrated in a physiological buffer solution. Results of these mucoadhesion studies as shown in Fig. 7 demonstrated that a certain fraction of applied particles was rapidly washed out. The rest, however, showed significant mucoadhesive properties. The mucoadhesion of particles comprising thiolated poly(acrylic acid) was significantly higher than that of particles being based on the corresponding unmodified polymer. Thiolated
Fig. 7. Adhesion of poly(acrylic acid) – cysteine microparticles ( ) and poly(acrylic acid) microparticles (o) on freshly excised small intestinal mucosa rinsed with 100 mM phosphate buffer, pH 6.0, at 37 jC as function of time. Indicated values are means ( F S.D.) of three to six experiments.
y
36
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
Fig. 8. Images of fluorescence-labelled poly(acrylic acid) – cysteine and poly(acrylic acid) microparticles on freshly excised small intestinal mucosa rinsed with 100 mM phosphate buffer, pH 6.0, at 37 jC a few minutes and 3 h after application. Left top: poly(acrylic acid) – cysteine microparticles, t = 5 min. Left bottom: poly(acrylic acid) – cysteine microparticles, t = 180 min. Right top: poly(acrylic acid) microparticles, t = 5 min. Right bottom: poly(acrylic acid) microparticles, t = 180 min.
when the adhesive bond fails within the mucoadhesive polymer itself [23].
4. Discussion Within this study, a model drug whose oral bioavailability should be improved by the use of mucoadhesive microparticles was used. The potential of orally given mucoadhesive microparticles
could meanwhile be verified by a study in human volunteers demonstrating a strongly improved oral bioavailability of riboflavin when the drug was incorporated in mucoadhesive microparticles [5]. Similar results might be achieved for bromelain being incorporated in poly(acrylic acid) – cysteine microparticles. Apart from an oral use of poly(acrylic acid) – cysteine microparticles, they might also be useful in order to improve the bioavailability of various poorly absorbed drugs being administered
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
via other transmucosal routes such as the nasal or vaginal mucosa. Vidgren et al. [24], for instance, could show in human volunteers a significantly prolonged residence time on the nasal mucosa for microparticles being based on unmodified poly(acrylic acid) in comparison to nonadhesive particles. As the mucoadhesive properties of poly(acrylic acid) microparticles could be strongly improved by the immobilization of thiol groups, a more pronounced effect as described by Vidgren et al. can be expected but has to be verified by according in vivo studies. Apart from the high cohesive and mucoadhesive properties, poly(acrylic acid) – cysteine microspheres should also lead to an improved drug uptake. Due to the intimate contact of adherent microparticles with the absorption membrane, a comparatively steep concentration gradient of the drug to the epithelium can be achieved, representing the driving force for a passive absorption. In addition, poly(acrylic acid) exhibits per se a permeation enhancing effect [25]. Moreover, Clausen and Bernkop-Schnu¨rch [26] could demonstrate that due to the immobilization of thiol groups on poly(acrylic acid), its permeation enhancing effect can be significantly improved. According to these results also, poly(acrylic acid) – cysteine microspheres should exhibit strong permeation enhancing properties. Recently, it was shown that this permeation enhancing effect of thiolated polymers can even be further improved by the addition of glutathione, which seems to mediate the permeation-enhancing effect of thiomers [27]. Hence, the addition of glutathione to poly(acrylic acid) – cysteine microspheres might be advantageous when an improved mucosal uptake of the incorporated drug is desired. In case of peptide and protein delivery, the enzyme inhibitory properties of poly(acrylic acid) should also contribute to an improved drug uptake. Poly(acrylic acid) displays a strong inhibitory effect towards the endopeptidase trypsin [28]. By the immobilization of thiol groups on poly(acrylic acid) also, a pronounced inhibitory effect towards exopeptidases such as carboxypeptidase A and B or aminopeptidase N can be achieved [29,30]. Being incorporated in poly(acrylic acid) – cysteine microparticles, therapeutic peptides or proteins such as bromelain should, therefore, be protected towards an enzymatic attack of these enzymes.
37
5. Conclusion Poly(acrylic acid) – cysteine microparticles display high cohesive properties due to a cross-linking process via the formation of disulfide bonds during preparation (I). Remaining free thiol groups seem to be responsible for relatively high mucoadhesive properties (II). Furthermore, a controlled drug release might be guaranteed by these particles (III) and— based on the results of prior studies performed with poly(acrylic acid) – cysteine—a permeation enhancing effect (IV) and enzyme inhibitory properties (V) can be expected. Because of these features, poly(acrylic acid) – cysteine microparticles seem to represent a promising novel delivery system for therapeutic agents whose uptake from the mucosal membranes such as the intestinal, nasal or vaginal should be prolonged and/or enhanced. Acknowledgements This work was supported by grant no. P13085MOB from the Fonds zur Fo¨rderung der wissenschaftlichen Forschung (FWF) to A. BernkopSchnu¨rch. The authors wish to thank Mr. Stro¨bel et al. from the slaughterhouse Totzenbach for supply of porcine intestinal mucosa. References [1] R.B. Shah, F. Ahsan, M.A. Khan, Oral delivery of proteins: progress and prognostication, Crit. Rev. Ther. Drug Carr. Syst. 19 (2002) 135 – 169. [2] A.A. Raoof, Z. Ramtoola, B. McKenna, R.Z. Yu, G. Hardee, R.S. Geary, Effect of sodium caprate on the intestinal absorption of two modified antisense oligonucleotides in pigs, Eur. J. Pharm. Sci. 17 (2002) 131 – 139. [3] S.R. Money, J.W. York, Development of oral heparin therapy for prophylaxis and treatment of deep venous thrombosis, Cardiovasc. Surg. 9 (2001) 211 – 218. [4] M.E. Klepser, M.N. Marangos, K.B. Patel, D.P. Nicolau, R. Quintiliani, C.H. Nightingale, Clinical pharmacokinetics of newer cephalosporins, Clin. Pharmacokinet. 28 (1995) 361 – 384. [5] Y. Akiyama, N. Nagahara, E. Nara, M. Kitano, S. Iwasa, I. Yamamoto, J. Azuma, Y. Ogawa, Evaluation of oral mucoadhesive microspheres in man on the basis of the pharmacokinetics of furosemide and riboflavin, compounds with limited gastrointestinal absorption sites, J. Pharm. Pharmacol. 50 (1998) 159 – 166.
38
A. Bernkop-Schnu¨rch et al. / Journal of Controlled Release 93 (2003) 29–38
[6] C.-M. Lehr, F.G.J. Poelma, H.E. Junginger, An estimate of turnover time of intestinal mucus gel layer in the rat in situ loop, Int. J. Pharm. 70 (1991) 235. [7] J.F. Forstner, Intestinal mucins in health and disease, Digestion 17 (1978) 234. [8] A. Allen, D.A. Hutton, J.P. Pearson, L.A. Sellers, in: J. Nugent, M. O’Connor (Eds.), Mucus and Mucosa, Ciba Foundation Symposium, vol. 109 (1998) p. 1984. [9] J.J. Sahlin, N.A. Peppas, Enhanced hydrogel adhesion by polymer interdiffusion: use of linear poly(ethylene glycol) as an adhesion promoter, J. Biomater. Sci., Polym. Ed. 8 (1997) 421 – 436. [10] B. Naisbett, J. Woodley, The potential use of tomato lectin for oral drug delivery. Lectin binding to rat small intestine in vitro, Int. J. Pharm. 107 (1994) 223 – 230. [11] A. Bernkop-Schnu¨rch, F. Gabor, M.P. Szostak, W. Lubitz, An adhesive drug delivery system based on K99-fimbriae, Eur. J. Pharm. Sci. 3 (1995) 293 – 299. [12] V.M. Leitner, G.F. Walker, A. Bernkop-Schnu¨rch, Thiolated polymers: evidence for the formation of disulphide bonds with mucus glycoproteins, Eur. J. Pharm. Biopharm. (2003) (in press). [13] J.D. Smart, I.W. Kellaway, H.E.C. Worthington, An in vitro investigation of mucosa-adhesive materials for use in controlled drug delivery, J. Pharm. Pharmacol. 36 (1984) 295 – 299. [14] M.K. Marschu¨tz, A. Bernkop-Schnu¨rch, Thiolated polymers: self-crosslinking properties of thiolated 450 kDa poly(acrylic acid) and their influence on mucoadhesion, Eur. J. Pharm. Sci. 15 (2002) 387 – 394. [15] A.J. Coupe, S.S. Davis, I.R. Wilding, Variation in gastrointestinal transit of pharmaceutical dosage forms in healthy subjects, Pharm. Res. 8 (1991) 360 – 364. [16] A. Bernkop-Schnu¨rch, C.E. Kast, M.F. Richter, Improvement in the mucoadhesive properties of alginate by the covalent attachment of cysteine, J. Control. Release 71 (2001) 277 – 285. [17] A.F. Habeeb, A sensitive method for localization of disulfide containing peptides in column effluents, Anal. Biochem. 56 (1973) 60 – 65. [18] K.V. Ranga Rao, P. Buri, A novel in situ method to test polymers and coated microparticles for bioadhesion, Int. J. Pharm. 52 (1989) 265 – 270. [19] B. Kriwet, E. Walter, T. Kissel, Synthesis of bioadhesive
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
poly(acrylic acid) nano- and microparticles using an inverse emulsion polymerization method for the entrapment of hydrophilic drug candidates, J. Control. Release 56 (1998) 149 – 158. A. Bernkop-Schnu¨rch, M.D. Hornof, C.E. Kast, N. Langoth, Thiolated polymers: stability of thiol moieties under different storage conditions, Sci. Pharm. 70 (2002) 331 – 339. A. Bernkop-Schnu¨rch, R. Giovanelli, C. Valenta, Peroral administration of enzymes: strategies to improve the galenic of dosage forms for trypsin and bromelain, Drug Dev. Ind. Pharm. 26 (2000) 115 – 121. C.-M. Lehr, From sticky stuff to sweet receptors—achievements, limits and novel approaches to bioadhesion, Eur. J. Drug Metab. Pharmacokinet. 21 (1996) 139 – 148. A. Bernkop-Schnu¨rch, S. Steininger, Synthesis and characterisation of mucoadhesive thiolated polymers, Int. J. Pharm. 194 (2000) 239 – 247. P. Vidgren, M. Vidgren, P. Vainio, J. Nuutinen, P. Paronen, Double-labelling technique in the evaluation of nasal mucoadhesion of disodium cromoglycate microspheres, Int. J. Pharm. 73 (1991) 131 – 136. C.E. Kast, A. Bernkop-Schnu¨rch, Influence of the molecular mass on the permeation enhancing effect of different poly(acrylates), STP Pharma 12 (2002) 351 – 356. A.E. Clausen, A. Bernkop-Schnu¨rch, In vitro evaluation of the permeation-enhancing effect of thiolated polycarbophil, J. Pharm. Sci. 89 (2000) 1253 – 1261. A.E. Clausen, C.E. Kast, A. Bernkop-Schnu¨rch, The role of glutathione in the permeation enhancing effect of thiolated polymers, Pharm. Res. 19 (2002) 602 – 608. H.L. Luessen, J.C. Verhoef, G. Borchard, C.-M. Lehr, A.G. de Boer, H.E. Junginger, Mucoadhesive polymers in peroral peptide drug delivery: II. Carbomer and polycarbophil are potent inhibitors of the intestinal proteolytic enzyme trypsin, Pharm. Res. 12 (1995) 1293 – 1298. A. Bernkop-Schnu¨rch, H. Zarti, G.F. Walker, Thiolation of polycarbophil enhances its inhibition of soluble and intestinal brush border membrane bound aminopeptidase N, J. Pharm. Sci. 90 (2001) 1907 – 1914. A. Bernkop-Schnu¨rch, S. Thaler, Polycarbophil – cysteine conjugates as platforms for oral (poly)peptide delivery systems, J. Pharm. Sci. 89 (2000) 901 – 909.